AbstractElectrocatalytic oxygen reduction reaction (ORR) has been intensively studied for environmentally benign applications. However, insufficient understanding of ORR 2 e⁻‐pathway mechanism at the atomic level inhibits rational design of catalysts with both high activity and selectivity, causing concerns including catalyst degradation due to Fenton reaction or poor efficiency of H₂O₂ electrosynthesis. Herein we show that the generally accepted ORR electrocatalyst design based on a Sabatier volcano plot argument optimises activity but is unable to account for the 2 e⁻‐pathway selectivity. Through electrochemical and operando spectroscopic studies on a series of CoN_x/carbon nanotube hybrids, a construction‐driven approach based on an extended “dynamic active site saturation” model that aims to create the maximum number of 2 e⁻ ORR sites by directing the secondary ORR electron transfer towards the 2 e⁻ intermediate is proven to be attainable by manipulating O₂ hydrogenation kinetics.